During the past few years studies by our group and others have established the importance of genetic factors in adult obesity and have found evidence for the action of major genes for obesity in some families. The goals of this study are to further clarify the mode of inheritance and to identify major genes conveying susceptibility to obesity in humans. Specifically, we propose to develop new family resources that will provide the types of information needed for quantitative and molecular genetic studies: 100 multiplex families with large sibships (with 689 total siblings, 344 of whom are obese), particularly those with a clinical risk factor for familial overweight such as childhood (before age 10) onset; 1 three-generation extended family (100 members) and 25 multiplex nuclear families (173 members) from a Pima Indian group having an extremely high prevalence of obesity; and 2 extended families (100 members) from an Amish population isolate having limited genetic and nongenetic variability. These samples are more likely to be segregating for obesity predisposing genes than are families identified in other ways. With these unique family resources, we propose: 1) to study alternative phenotypes for obesity - overweight indices, body composition, fat distribution and metabolic rate, 2) to further clarify the mode of inheritance of obesity and related phenotypes through segregation analysis and analyses of the effects of childhood onset and birth cohort on individual and familial risk for obesity, and 3) to use non-parametric (sibling and affected family member pairs) and parametric (lod score) linkage methods to identify major genes for obesity. The structures, sizes and sources of these families combined with our approaches to phenotype definition and quantitative and molecular analyses will provide us with the power needed to achieve our goals. Besides the obvious importance of obesity to physical and mental health, obesity can serve as a model for the genetic study of many complex human disorders.